Page 47 - Handbook of Biomechatronics

Actuator Technologies 41

density). For biomechatronic applications in which acceleration occurs, the
speed ratio is a more relevant compact notation, and motor envelope visu-
alizations are even better, at the expense of being less compact a comparison.
Weight calculation (as a function of torque densities, and including electrical
energy density necessary to accomplish the task), although more involved, is
a powerful metric that has enabled significant improvements in the field.

2.3 Ease of Use
How do I make this actuator easy for humans to use?
Potentially the greatest challenge of biomechatronics is designing a sys-
tem that is intuitive and comfortable. Actuators play an important role in
achieving this challenge: they should generate movement in a way that peo-
ple can learn to predict. Actuators with hard nonlinearities (e.g., stiction and
backlash) feel unnatural to the human user. Actuators with soft nonlinearities
(e.g., quadratic viscous drag) are fine as long as the human user is able to
control and predict them.
One strategy to test ease of use is to study the closed-loop performance of
the human and the mechatronic device together. Testing with the human
operator is of course the gold standard; however, the challenge is that a
device needs to be designed and functional before testing is possible. Often-
times, modifying actuator parameters may require complete redesigns of
hardware systems. To overcome these challenges and decrease the time
required for design iterations, a flexible testing platform can be very helpful.
For example, one group developed a prosthesis emulator that allows testing
of a wide range of prosthesis parameters without the need to design multiple
devices (Caputo and Collins, 2014).
Another strategy is to study human perception and control, and use that
knowledge to choose system characteristics. Psychophysics methods can be
used to quantify how precisely human users are able to predict movements
with the biomechatronic system ( Johnson et al., 2017). System identifica-
tion methods can be used to quantify typical human body parameters such
as mechanical joint impedance (Rouse et al., 2014), which can then be mim-
icked by biomechatronic actuators.

3 TYPES OF BIOMECHATRONIC ACTUATORS

There have been several helpful categorizations of actuators that drive
biomechatronic systems. Hannaford and Winters categorized actuators by